In the
Court of Appeals
Second Appellate District of Texas
at Fort Worth
No. 02-21-00174-CR
No. 02-21-00175-CR
No. 02-21-00176-CR
EMANUEL OCHOA, Appellant
V.
THE STATE OF TEXAS
On Appeal from the 235th District Court
Cooke County, Texas
Trial Court Nos. CR19-00054, CR19-00056, CR19-00057
Before Birdwell, Womack, and Wallach, JJ.
Opinion on Remand by Justice Wallach
OPINION ON REMAND
The jury convicted Appellant Emanuel Ochoa of aggravated kidnapping, injury
to a child, and aggravated sexual assault of a child under six. The evidence presented
to the jury included the video recording of a statement that Ochoa had given to Texas
Ranger James Holland in which Ochoa confessed to removing five-year-old M.G. 1
from her bed in the middle of the night, taking her to the nearby empty trailer where
he used to live, sexually assaulting her, punching her in the head to “put her
unconscious,” and then leaving her in freezing temperatures underneath another
empty trailer nearby. Ochoa was fourteen at the time. Ochoa’s confession was the
strongest, clearest, and most persuasive evidence of guilt that the State presented at
trial.
Holding the confession admissible, this court affirmed his conviction. Ochoa v.
State, 675 S.W.3d 793, 814 (Tex. App.—Fort Worth 2023), rev’d and remanded,
707 S.W.3d 344 (Tex. Crim. App. 2024). Ochoa petitioned for discretionary review
with the Court of Criminal Appeals, which was refused. But the court granted review
on its own motion to consider whether Ochoa’s confession was involuntary. The
court ultimately held that Ochoa’s statement was inadmissible because it had not been
made voluntarily, and it remanded the case for this court to conduct a harm analysis.
We use initials for the child to protect her privacy. See Tex. R. App. P. 9.8 cmt.,
1
9.10(a)(3); McClendon v. State, 643 S.W.2d 936, 936 n.1 (Tex. Crim. App. [Panel Op.] 1982).
2
Ochoa v. State, 707 S.W.3d 344, 366 (Tex. Crim. App. 2024). Because we reach the
almost inescapable conclusion that the admission of Ochoa’s statement caused harm
under the legally-required standard, we will reverse the trial court’s judgment and
remand the case to the trial court.
Background
This court’s prior opinion discussed the events leading to the criminal
investigation in this case, and thus we recite here only the facts essential to our review.
M.G. lived in a trailer 2 with twelve other people: her brother, her father and his
girlfriend, Ochoa and his mother, a man named Jeremiah Jacques, Jacques’s girlfriend,
and others. When M.G.’s disappearance was discovered on the morning of February
6, 2018, law enforcement was contacted and a search was conducted, but M.G. was
not located until mid-afternoon when she was found by Ochoa and Jacques under a
trailer two doors down from their home. She was wearing a nightgown but no
underwear, and she had a black trash bag around the lower half of her body. She had
been sexually assaulted and had suffered an injury that resulted in permanent brain
damage.
2
The trial witnesses used the term “trailer,” so we use that term here. Because the term “trailer” can refer to different types of accommodations, we specify that the home was what is commonly referred to as manufactured housing. Regardless of the term, from the testimony and photographs admitted at trial, the home did not appear to comfortably accommodate the number of people living there.
3
Law enforcement officers decided to start their investigation by interviewing
the two people who had found her—Jacques and Ochoa—because the investigators
“wanted to start right there as a baseline.” Investigators initially were suspicious of
Jacques because they had found his emotional reactions to her disappearance to be
odd. However, the two Texas Rangers who interviewed him ruled him out as a
primary suspect based on this interview.3 Ranger Holland then interviewed Ochoa,
and in that interview Ochoa confessed and said that he had taken M.G. to the trailer
immediately next door where he had once lived,4 assaulted her, cleaned her up, and
then left her underneath the other trailer, where he believed she would die. He denied
wrapping the black trash bag around her.
At trial, the court admitted Ochoa’s confession, and Ranger Holland testified
about what Ochoa had told him and his impression and observations of Ochoa
during the interview. As we discuss more below, the trial court also admitted
testimony about the results of DNA tests conducted on various items, including
Ochoa’s underwear.
3
Ranger Holland testified that he “felt like [Jacques’s] responses, although they were weird at times, they would be fairly normal for his personality.” Ranger Holland did not see in Jacques “specific indications of what [he] thought [he] would see from the person who did this,” and Jacques “was able to answer questions[,] and what he said at the end of the day made sense.”
4
This trailer was located between the trailer where they lived and the one under which she was found.
4
Harm Standard
When an involuntary statement is improperly admitted, we conduct a harm
analysis to determine whether the error requires reversing the judgment. Kane v. State,
173 S.W.3d 589, 594 (Tex. App.—Fort Worth 2005, no pet.). Because the error is
constitutional, we apply Rule of Appellate Procedure 44.2(a) and must reverse unless
we determine beyond a reasonable doubt that the error did not contribute to
appellant’s conviction or punishment. Id. (citing Tex. R. App. P. 44.2(a)). If there is a
reasonable likelihood that the error affected the jury’s deliberations, the error is not
harmless beyond a reasonable doubt. Wesbrook v. State, 29 S.W.3d 103, 119 (Tex. Crim.
App. 2000); see also Neal v. State, 256 S.W.3d 264, 284 (Tex. Crim. App. 2008).
Our harm analysis should not focus on the propriety of the outcome of the
trial; instead, we focus on whether the constitutional error adversely affected the
integrity of the process leading to the conviction. See Wells v. State, 611 S.W.3d 396,
410 (Tex. Crim. App. 2020); see also Wesbrook, 29 S.W.3d at 119 (stating that the
appellate court “should calculate as much as possible” the error’s probable effect on
the jury “in light of the existence of other evidence”). To that end, in evaluating harm,
we should consider “any and every circumstance apparent in the record that logically
informs” our analysis. Snowden v. State, 353 S.W.3d 815, 822 (Tex. Crim. App. 2011)
(quoting Tex. R. App. P. 44.2(a)).
The most significant concern is the error and its effects, Staley v. State,
No. 02-23-00053-CR, 2025 WL 727842, at *16 (Tex. App.—Fort Worth Mar. 6, 2025,
5
pet. granted) (citing Wesbrook, 29 S.W.3d at 119), and “a constitutional harm analysis
does not turn on whether, discounting the erroneously admitted evidence, the
remaining evidence was legally sufficient to convict,” Scott v. State, 227 S.W.3d 670,
694 (Tex. Crim. App. 2007). However, the presence of overwhelming evidence
supporting the jury’s finding can be a factor in evaluating harmless error. Wells,
611 S.W.3d at 410. Other factors to consider may include, if applicable, the nature of
the error, the extent that the State emphasized it, its probable collateral implications,
and how a juror would likely have weighed it “in the course of [jury] deliberations.” Id.
at 410. The entire record is to be evaluated in a neutral manner and not in the light
most favorable to the prosecution. Id. at 410–11.
Discussion
In his sole point on remand, Ochoa argues that he was harmed by his
confession’s admission. See Carroll v. State, 101 S.W.3d 454, 460–61 (Tex. Crim. App.
2003) (discussing issues appellant may raise on remand). We begin our analysis by
recognizing that a defendant’s confession has a profound effect on the jury and is
“‘probably the most [probative and] damaging evidence that can be admitted’” to
establish the defendant’s guilt. Pecina v. State, 326 S.W.3d 249, 268–69 (Tex. App.—
Fort Worth 2010) (quoting Arizona v. Fulminante, 499 U.S. 279, 292, 111 S. Ct. 1246,
1255 (1991) (White, J., dissenting)), rev’d on other grounds by, 361 S.W.3d 68 (Tex. Crim.
App. 2012). Indeed, if a jury knows that a defendant has confessed, “‘it will doubtless
be tempted to rest its decision on that evidence alone, without careful consideration
6
of the other evidence in the case.’” Id. at 269 (quoting Fulminante, 499 U.S. at 313,
111 S. Ct. at 1266 (Kennedy, J., concurring)). With those admonitions in mind, based
on our review of the entire record, we hold that the record establishes harm.
I. Opening Statements
The State did not spend long discussing Ochoa’s confession during its opening
statement; the prosecutor’s opening statement takes up approximately ten pages of
the reporter’s record, most of which focuses on the investigation up until Ochoa’s
confession. The discussion of the statement takes up around three paragraphs. But
those three paragraphs emphasized Ochoa’s confession by telling the jury what
happened to M.G. through Ochoa’s description of the event:
[A]t first, he denies it, denies it, denies it, but finally, he tells Ranger
[James] Holland exactly what happened.
He says that he went into her room, at about 5:00 that morning,
put a blanket over her ahead so that she wouldn’t know who it was,
brought her from that room where she was sleeping, and took her to the
trailer that he used to live in. He had been evicted from the trailer next
door, and h[e] and his family moved into this trailer with another family.
He brought her to the trailer that they had been evicted from, and
he tells Ranger Holland that—that he raped her. He tells Ranger Holland
that he put his penis in her vagina and that he had sex with her.
And he says that during this, she was screaming and crying and
fighting him, hitting him. And at some point, she wouldn’t be quiet, so he
tells Ranger Holland that he hit her in the back of the head. Another
point during the interview, he explains that he put her to sleep. And then
he explains to Ranger Holland that he picked her up, brought her to under
the trailer, and covered her with her blanket and went back to bed.
[Emphasis added.]
7
The prosecutor then told the jury that it would hear other evidence as well,
including evidence that M.G.’s DNA had been found on the inside fly of Ochoa’s
underwear. Although the prosecutor did not discuss the confession at length, in those
three paragraphs the prosecutor provided the jury with a filter through which to hear
the evidence. Ochoa’s attorney also addressed the confession in his opening
statement, telling the jury that Ranger Holland had manipulated Ochoa.
II. Trial Evidence 5
A. DNA Profiles from Swabs of M.G. and Ochoa’s Underwear
The State’s brief relies heavily on DNA evidence admitted at trial. We therefore
discuss that testimony in detail and explain why, although the evidence does not
establish Ochoa’s innocence, it is far from being such strong evidence of guilt that it
can overcome the harm from the confession’s admission.
Rachel Burch, a forensic analyst at University of North Texas Center for
Human Identification testified about how DNA testing works and the analysis that
she had performed in this case. Burch first explained how DNA is extracted,
quantified, amplified, and then analyzed:
So what we’re -- we’ll do is an extraction step, so what that’s doing is it’s
taking these cells that are on that swab [or other item being tested], and
causing them to open up and release their DNA. Then we do a step that
is a quantification step, and it lets us know how much DNA we got from
5
We omit discussion of the evidence that showed that the offenses had been committed but did not point toward any particular culprit, such as the evidence relating to the extent of M.G.’s injuries.
8
the extraction. Then we do an amplification step, so what that’s going to
do is it’s going to make copies of the DNA in those areas that we’re
interested in. And then we’re going to take that copied or amplified
DNA, and put it on an instrument that will allow us to visually see a
person’s DNA profile.
She also explained how a person is included or excluded as a contributor to a
DNA profile found in that process:
And in forensics, we’re interested in areas of DNA that differ between
individuals. And what we’re actually looking for are these little sections
of DNA that repeat over and over again, and we count the number of
repeats a person gets from their mother and the number of repeats a
person gets from their father. And when we compile all these repeats
together, we have a person’s DNA profile.
So what we’ll do is we’ll take the repeat types that we detected in
the evidence and compare it to the repeat types we have for a person’s
known DNA profile. And if all of the types are present within the
evidence profile, then we say somebody cannot be excluded as a possible
contributor, which would mean that they’re included as a possible
contributor, to the DNA that we’re detecting. If the types aren’t present,
then a person is excluded as being a possible contributor.
So if we have a mixture[, meaning] DNA from more than one
person, sometimes we don’t know [how] many people are even in that
mixture, so we’re not going to feel comfortable saying how many people
are in the mixture and who could be included or who could be excluded,
so we’re going to consider the data or a comparison to it as inconclusive
For somebody to be included, we would have to see their types at
every location[, w]ith the exception of a location where we thought that
that type may not be present[.] [B]ut we have dictated that up front
9
based off of the data, so we would already have noted that before we
make a comparison to the known.
Burke then discussed the testing performed on Ochoa’s underwear and swabs
taken from M.G. by the sexual assault nurse examiner.
1. Inner Labia Swab
In discussing two swabs taken from M.G.’s labia in the sexual assault
examination, Burke explained three different steps that an analyst may take to look for
the presence of seminal fluid. The first test detects acid phosphatase. Although this
chemical is produced by men and women, “males have more of it in their seminal
fluids,” so if the test is positive, that “lets [the analyst] know . . . to look a little bit
further into that sample.”
The second step is a “prostate specific antigen test” that detects a protein made
by the prostate. Burke stated that “seminal fluid, male urine, [and] male blood” will
“give positive ‘p30’ results.” If that test is positive, “that lets [the analyst] know that
[they] might need to still keep looking at that sample.” But she also said that this
antigen can also be—though more rarely—found in vaginal secretions and “some
other things.”
The third step is “tak[ing] a portion of the swab and get[ting] the cells off of it
and put that on a microscope slide and physically look for the presence of sperm cells.
If [the analyst] see[s] a sperm cell or sperm cells, then that’s the way that [the analyst]
can definitely say that [the swab has] seminal fluid present or semen.”
10
Burke applied these tests to swabs taken from M.G.’s inner and outer labia.
The swab taken from the inner labia was positive on the first two presumptive tests,
and when the sample was examined under a microscope, Burke saw one sperm cell.
Thus, the testing confirmed the presence of semen. However, when the lab attempted
to “do DNA testing that is specific to males,” they “had no Y-STR[6] results for that,”
so they were “not able to make a comparison to anything.”
2. Outer Labia Swab
As for the outer labia swab, it was negative on the acid phosphatase test but
positive on the p30 test. However, no sperm cells were detected on the microscope
slide, so the lab was unable to confirm the presence of semen. The sample was then
subjected to testing for a DNA profile. The lab report briefly described the two types
of extractions performed: “Differential DNA extractions . . . result[ ] in two fractions,
an epithelial fraction and a sperm fraction. The epithelial fraction is expected to
primarily contain non-sperm cell DNA and the sperm fraction is expected to primarily
contain sperm cell DNA.”
Burke provided the jury with a more detailed explanation. She first described
the process to extract a sample’s epithelial fraction, which involves opening up any
cells that are not sperm cells:
Burke did not explain to the jury what the terms “STR” or “Y-STR” refer to,
6
but she had explained that an analyst looks for “little sections of DNA that repeat over and over again.” See National Institute of Justice, “What Is STR Analysis?,” March 2, 2011, https://nij.ojp.gov/topics/articles/what-str-analysis.
11
[W]henever we have semen screening results that indicate that sperm
might be present, we do a special extraction, and that special extraction
takes advantage of the fact that sperm cells, if they’re present, are
actually more resilient so they won’t break open as easily as . . . skin cells
will, and so we’ll add a set of chemicals that will cause the cells to open
up that are not sperm cells. We call that the epithelial fraction. So the
first result is for that first fraction, that epithelial fraction.
Once the process has opened up the cells that are not sperm cells—i.e., the
epithelial fraction of the sample—that DNA is removed from the sample, leaving
behind the sperm cells, if any are present. If the sample contains sperm cells, those
cells remain in the sample. This remaining part of the sample is referred to as the
sperm-cell fraction. The lab applies “a different set of chemicals that will cause sperm
cells, if they’re present, to open up. And then [the lab] do[es] a DNA profile on that.”
[Emphasis added.] In other words, the process allows the lab to first try to detect a
DNA profile in the epithelial fraction and then, if there are any sperm cells present, in
the sperm-cell fraction.
In the sample taken from the swab of M.G.’s outer labia, the p30 test was
positive, but no sperm cells were detected on the microscope slide, and thus the
preliminary testing could not confirm the presence of semen. However, because the
p30 test was positive, the lab processed the sample to separate out the epithelial
fraction from any sperm cells that might have been present; the epithelial fraction
12
yielded “[i]nconclusive Y-STR results,”7 and only a partial STR profile was obtained
from the sperm-cell fraction. M.G. could not be excluded as a contributor to that
partial profile in the sperm-cell fraction. Burke explained how M.G. could not be
excluded as the contributor even though “females don’t have sperm cells”: “[W]hen
we added those chemicals and opened up the epithelial cells, when we remove that,
we try to wash away that DNA before we open up any sperm cells if they’re present,
so this is just leftover female DNA [from the epithelial cells] that we couldn’t wash
away” from the sperm cell fraction. While M.G. could not be excluded, Ochoa was
excluded as a contributor. In summary, this testing provided no evidence of Ochoa’s
guilt. 8
3. Other Swabs
The sexual assault nurse examiner took other swabs of M.G., but the presence
of semen could not be confirmed on any of those samples. Burke did not testify about
further testing of these samples, but the lab report states that for one set of swabs,
Burke did not provide any details about why the epithelial fraction of this
7
sample was deemed inconclusive, such as whether it was inconclusive as to the presence of male DNA or inconclusive as to the profile of the contributor.
According to Burke, the testing could not confirm that there was any semen
8
on M.G.’s outer labia because there were no sperm cells detected, and although the test had a positive p30 result, Burke had testified that it is possible for a positive p30 to come from a female source. Thus, Burke’s testimony could have appeared to the jury to leave open the possibility that there had been no male contributor to this sample. But regardless of whether the testing of this sample could be said to implicate another person, it clearly did not point specifically to Ochoa.
13
“[n]o Y-STR results were obtained from the epithelial and sperm fractions.” In other
words, lab testing on these swabs did not implicate Ochoa or any other specific
person.
4. Ochoa’s Underwear
As for the testing on Ochoa’s underwear, the State presents the results as
strong evidence supporting Ochoa’s guilt. The lab tested three sections that were
described in the report as the outer front fly area, a stained area on the outside of the
underwear near the waistband, and a stained area on the inner front fly.
a. Stained Area Near Waistband
The lab performed presumptive tests for blood on the inner front fly area and
the stained area near the waistband. That test detects hemoglobin, a component in the
blood of humans, higher primates, and ferrets. The test of the area near the waistband
was positive for hemoglobin. Testing also revealed the presence of semen. However,
the lab did not attempt to extract a DNA profile from the sample.
b. Inner Front Fly Area
The inner front fly area also tested positive for hemoglobin. Testing detected
the presence of semen; the presumptive tests were positive and sperm cells were
identified in the sample. The lab then performed the differential extraction process
discussed above, and the epithelial fraction revealed a mixed DNA profile “with a
major male contributor and at least one minor contributor.” Ochoa could not be
excluded as the major contributor—not a surprising fact, since the underwear was
14
that of a fourteen-year-old boy. More importantly, assuming there was one minor
contributor, M.G. could not be excluded as that minor contributor. The State
emphasizes this DNA evidence in its brief.
However, despite the lab report’s use of “outer” and “inner” to describe the
part of the underwear being tested, Burke testified that the lab had taken a cutting to
perform the testing on that area, and “the cutting goes through the front in—all the
way through, so the cutting will—the DNA from that could be from the outside or
the inside.” It is not clear whether Burke meant that the act of taking the cutting could
contaminate one side of the cutting with DNA from the other side, that the testing
process would pull DNA from both sides of the material, or something else.
Regardless, from Burke’s testimony, M.G.’s DNA could have been located on the
outside of the underwear rather than the inside.
Ochoa and M.G. lived together in a small, crowded home, and there was no
testimony ruling out a non-incriminating explanation for the DNA’s presence on the
underwear. To the contrary, Burke testified that she could not say whether M.G.’s
DNA had come from the blood or skin cells or some other source, and she described
to the jury how DNA can transfer between people by a person touching an object
that another person then touches.
15
c. Stained Outer Fly Area
Testing confirmed the presence of semen on this area. Burke did not testify
about any further testing done on this item, and the lab report did not discuss any
further testing.
B. Other Evidence
On the first day of trial, the jury heard testimony from a paramedic who had
transported M.G. to the hospital. The paramedic testified that M.G. had said “that
they were playing hide and seek and that ‘Hakea’ and ‘Davine’ put her under the
house when it was dark.” The paramedic had a hard time hearing M.G., who was
soft-spoken, so the paramedic “couldn’t tell if it was two people or she was using two
words” and could not guarantee that she had heard either word correctly. At one
point, M.G. told the paramedic that “she didn’t know why he was on top of her,” but
she did not say who “he” was.
Ranger Holland testified on the second day of trial. Most of his testimony
concerned Ochoa’s confession and the elicitation of it, including testimony related to
whether the confession had been made voluntarily. The magistrate who had provided
Ochoa with warnings required by Family Code Section 51.095 testified about his
administration of the warnings. See Tex. Fam. Code § 51.095. Further, the State played
for the jury the video recording of Ochoa’s confession, and the evidentiary strength of
the confession cannot be overlooked. As we noted above, the confession was the
strongest, clearest, most persuasive evidence that the State had at trial. It was the only
16
source for the specific details that the jury heard about the commission of the
offenses. Further, the video of the confession was particularly compelling because it
showed the process of Ranger Holland eliciting the confession from Ochoa—his
telling Ochoa about his extensive law enforcement experience, saying that he knew
that Ochoa was lying, telling Ochoa that M.G. was talking about what had happened
to her, and assuring Ochoa that he could help if Ochoa told the truth—and Ochoa’s
eventually describing how he had taken M.G. to the nearby trailer and raped her. The
video conveyed those details to the jury far more effectively and carried more
emotional weight than if Ranger Holland had simply described the confession for the
jury.
The State presented some evidence confirming Ochoa’s confession that the
assault had taken place in the trailer where Ochoa had once lived. For example, the
sexual assault nurse examiner who examined M.G. at the hospital testified that M.G.
smelled like cleaning solution, and a deputy who had ridden in the ambulance with her
said that he had noticed the smell of “some type of cleaning solvent” in the
ambulance. Ranger Holland also noticed the smell in the ambulance. Several law
enforcement witnesses testified that when they searched the trailer, they noticed a
strong odor of some sort of cleaning chemical. Further, the jury had some evidence
that M.G. was found with oatmeal on her skin, and the jury heard from other
witnesses that when the trailer was searched by law enforcement, they found oatmeal
on the floor. Moreover, a law enforcement witness testified that the trailer contained
17
multiple items that could have been used to make the ligature marks on M.G.’s neck.
Thus, the jury had some evidence aside from Ochoa’s confession that the trailer was
where M.G. had been assaulted.
The State then presented some evidence aside from Ochoa’s confession that
tied him to the trailer. The jury heard testimony that Ochoa used to live in the trailer.
Ranger Holland testified that he smelled a cleaning solution smell in the interview
room where he interviewed Ochoa, and he indicated that the smell seemed to be
coming from Ochoa. Ranger Holland had not noticed the smell when he interviewed
Jacques, even though an investigator had testified that Jacques had been the person
who had pulled M.G. out from underneath the trailer.
C. Testimony about Persons Not Investigated
The jury also heard evidence of what the investigators had not done in this case.
For example, Jerry Crumley, the lead investigator on the case, did not recall any
interview of M.G.’s father by law enforcement after she was located. Yet two
investigators from the Department of Family and Protective Services testified about
the conversation each had with M.G’s father, one at the scene before she was found
and the other several days later. They both testified that in their conversation, the
father acted as though he were crying but did not appear to be actually upset. The
investigator who conducted the second interview testified that the father “would
cover his face and make crying sounds, but then would like peek above his fingers to
look to see if we were looking at him. And when he would pull his hands down, there
18
was no tears, he did not appear to be in a state of crying.” Law enforcement did not
send a sample from the father to the forensic lab for testing.
Law enforcement also did not send any samples for testing from Jacques, even
though he also lived in the home with M.G. and several witnesses testified that he had
acted strangely, or for that matter from anyone else who lived in the home other than
M.G. and Ochoa. 9 Additionally, Crumley testified that in 2021, he had been asked to
resign or be terminated from his employment with the Cooke County Sheriff’s Office
because “the District Attorney’s Office would no longer accept [his] cases or [his]
testimony.”
In retrospect, a reasonable inference that could be drawn is that once they had
their man identified by the confession, the search for other potential culprits was
abandoned.
D. Evidence Summary
Some evidence linked M.G. and Ochoa to a nearby trailer where the assault
could have occurred. The results of testing on the swab of M.G.’s inner labia
supported a finding that M.G. had been sexually assaulted, but the test results did not
specifically implicate Ochoa or anyone else. The outer labia swab also failed to
implicate Ochoa. The testing of Ochoa’s underwear indicated the presence of blood,
and M.G. could not be excluded as a contributor to a DNA profile found on the
9
Crumley explained that this was because the lab “told [the investigators] that [they] had to choose a certain number of items.”
19
underwear. However, Burke testified that she could not say whether the DNA had
been on the inside of the underwear, she acknowledged that DNA can innocently
transfer to a person from an object, and other evidence established that Ochoa and
M.G. lived together in a small, overcrowded home. Additionally, the jury heard that
law enforcement had not investigated M.G.’s father or Jacques or sent any samples
from them to the lab for testing, despite both acting strangely. If this evidence raised a
doubt in the jurors’ minds about the investigation’s accuracy or thoroughness, the
confession gave them a reason to set those doubts aside and convict Ochoa anyway.
In sum, even if the trial evidence as a whole would have been sufficient to convict
Ochoa in the absence of the confession, it was not so overwhelming as to render the
confession’s admission harmless.
III. Closing Arguments and Jury Notes
In its closing arguments, the State argued that even without Ochoa’s
confession, the jury had sufficient evidence to find Ochoa guilty. But the
prosecutor—and in response, the defense—discussed the confession at length. Much
of this argument concerned whether the jury should disregard Ochoa’s confession on
the ground that it had been illegally obtained. During that argument, the prosecutor
told the jury that it should consider Ochoa’s confession and repeatedly stated that it
was true:
[Ranger Holland] did his job and they’re going to tell you, because he is
good at his job and got the truth in this case, the truth that was
20
corroborated by all of the evidence, the truth that this happened in his old
house, the truth that Emanuel Ochoa raped her.
Those things are things that we can prove happened by hard
evidence. And they’re going to tell you, ignore all of that because Ranger
Holland is good at his job and he got the truth. That’s what they’re going
to tell you.
But, ladies and gentlemen, you don’t have to ignore any of that
evidence. [Emphasis added.]
The defense closing arguments take up approximately thirty pages of the
reporter’s record, and about a third of that time was spent arguing that the jury should
disregard the confession. In rebuttal, the prosecutor spent a large part of his time
discussing other evidence—the cleaning solution smell, the fact that oatmeal had been
found in the trailer and on M.G., and the DNA evidence—and discounting what the
defense had characterized as failures in the investigation, like not pursuing other
suspects after Ochoa’s confession. The prosecutor did not spend much time
emphasizing the evidentiary value of the confession, but he did spend about four
pages of the record (out of the fifteen pages comprising his rebuttal) arguing that the
confession had not been illegally obtained. He told the jury, “[T]his is just the only
thing that they have to hold on to is for you to somehow ignore that confession. And
you don’t have to ignore it.” In short, although the prosecutor did not emphasize the
confession more than other evidence in his closing, neither did he minimize it, and he
strongly argued that the jury should consider it.
21
As the State notes in its brief, during deliberations, the jury asked to see the lab
report with the DNA test results. Although this request is some indication that the
jury considered all the evidence, not just Ochoa’s confession, we cannot say from this
request that the jury put more importance on the DNA testing, and we cannot
conclude that the DNA testing rendered the confession harmless.
IV. Conclusion: the Confession’s Admission Was Harmful
In summary, we conclude that the confession’s admission was harmful.
Although the State’s brief focuses on the DNA evidence, as noted above, this court’s
role is not to determine whether the DNA evidence could support the jury’s guilty
verdict but rather to calculate the probable effect that the confession had on the jury.
See Butler v. State, No. 02-17-00171-CR, 2018 WL 2343653, at *3 (Tex. App.—Fort
Worth May 24, 2018, no pet.) (mem. op., not designated for publication). Ochoa’s
confession was clear, detailed, and compelling to watch, and the video of it was played
for the jury, it was the primary point of Ranger Holland’s testimony, and the State
focused on it repeatedly throughout the trial. Moreover, the confession affected the
entire trial from start to end, and it allowed the jury to overlook any failures in the
investigation of the case. The other evidence of guilt in this case was not so
overwhelming that it could overcome the probable effect of the confession. Wesbrook,
29 S.W.3d at 119 (noting that overwhelming evidence supporting a finding can be a
factor in evaluating harm). We cannot determine beyond a reasonable doubt that the
22
confession made no contribution to Ochoa’s conviction, and we therefore sustain his
point on remand.
Conclusion
Having sustained Ochoa’s sole point, we reverse the trial court’s judgment and
remand this case for further proceedings consistent with this opinion. See Pecina,
326 S.W.3d at 271 (remanding for new trial after holding that defendant was harmed
by admission of his statement in violation of constitutional rights).
/s/ Mike Wallach
Mike Wallach
Justice
Publish
Delivered: July 9, 2026
23